Tillämpning av statistisk försöksplanering vid industriell brödtillverkning
Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
In a world where quality is increasingly important, the industry has a need to continuously improve product quality. One way to improve quality is through the use of statistical methods which can produce information needed to better understand cause and effect relationships in production processes. Several of these statistical methods, such as design of experiments and statistical process control, are however not used in a large scale in industry due to usage barriers. This thesis aims to lower one of these barriers which is the lack of a tangible approach to design of experiments, which is one of the least used statistical methods in the industry. This thesis consists of a case study at a company within the food industry where several cycles of design of experiments are executed in order to identify improvements to the design of experiments process as well as improving product quality. A bread product is chosen to be investigated. The case study includes analysis of historical production data and previous studies as well as interviews with employees in order to identify possible affecting factors. Based on this screening, three cycles of design of experiments are performed. The mixing time of the dough is investigated in the first cycle. The second cycle includes the mixing time, the fermentation time of the bread as well as the fermentation temperature. The last experiment once again focuses on the mixing time of the dough. The result shows that the mixing time does not affect the bread quality but the fermenting time and temperature affects the height of the products.While performing the design of experiments process several complications and differences from the recommended approaches in the literature occurred. For example this thesis uses historical data, interviews and previous studies as means for screening while the literature typically recommends the use of fractional factorial experiments. Based on these complications and differences the recommendations of the thesis are constructed to improve the design of experiment process.The recommended design of experiments approach includes includes six phases. One of the phases is newly developed by the author and the purpose is to make use of the production operators and their experiences from performing the experiments. The contents of the phases are also developed from the case study where a new checklist is made for the planning of future experiments. The checklist is based on the literature as well as conclusions from this study and it aims to address complications that occurred in this thesis.
Place, publisher, year, edition, pages
2015. , 75 p.
Social Behaviour Law
Samhälls-, beteendevetenskap, juridik, Statistisk försöksplanering, försöksplanering, processoptimering, kvalitet, brödtillverkning, design of experiments, DOE, process optimization, quality, bread manufacturing
IdentifiersURN: urn:nbn:se:ltu:diva-53312Local ID: a58fb3fc-c3a8-4fd3-babb-739c2374b6e4OAI: oai:DiVA.org:ltu-53312DiVA: diva2:1026686
Subject / course
Student thesis, at least 30 credits
Industrial and Management Engineering, master's level
Validerat; 20150814 (global_studentproject_submitter)2016-10-042016-10-04Bibliographically approved